Wind power prediction using recurrent multilayer perceptron neural networks

The power generated by the wind changes rapidly because of the continuous fluctuation of wind speed and direction. It is important for the power industry to have the capability to predict the power produced by the wind for the power management and control. In this paper, temporal characteristics of wind power generation are studied and recurrent multilayer perceptron (RMLP) neural networks are used to predict the power. The extended Kalman filter based backpropagation through time algorithm is used to train the RMLP networks. The paper demonstrates the RMLP network solution for the power prediction, and show that the RMLPs can be used to predict the wind power in changing wind conditions.